A Case Based Reasoning System For Diagnosis Of Malnutrition For Under-Five Year Children

Abstract

Malnutrition is an umbrella term for the unsuitable intake of nutrients needed to sustain healthy development; the term applies to under or over nutrition. It is possibly known to be one of the predominant causes of illness and death for under-five children in Ethiopia. Some of the reasons which worsen the spread of malnutrition in the country are: lack of specialists, practitioners and health facilities at lower level health institutions in order to diagnose and give treatment at early stage. Artificial Intelligence, which uses computer applications by simulating human intelligence, was applied in the research especially for malnutrition diagnosis. The general objective of this study was to design a case based reasoning system that provides expert advice for diagnosis of malnutrition under five year children. The cases were collected from both Jimma University specialized hospital and Hawasa university comprehensive specialized hospitals and design science were followed to design prototype case based reasoning system. Stratified sampling technique was employed to select domain experts for knowledge acquisition and for system testing and evaluation from Jimma University specialized hospital. For the development of the prototype system, the researcher used jCOLIBRI version 1.1 implementation tools and nearest neighbor algorithm. Evaluation of the developed prototype was performed for both system performance and user acceptance. For testing of the prototype seven test cases and six domain experts were used. Based on evaluating the performance of the system, the average precision and recall values achieved were 71% and 83% respectively. User acceptance testing also performed by involving domain experts and an average of 83% acceptance was achieved. Insertion of additional cases could increase the performance of the CBR system. In this study a promising result was obtained and met the objectives of the study.